Why the result is better than that in the papers? #87

[Nicozwy]

Hi,
I run the code on the conll2003 dataset, however, the result of the code is better than that reported in the paper. I am curious about where to get the basic version code of the paper, and how to improve the result?

Thank you

[ivsanro1]

Probably because paper is using Senna embeddings while this repository is using GloVe embeddings